People with strong political views about immigration can wield significant influence by crafting tweets laden with metaphors, shaping how others grasp this hotly contested issue.
In a new study, researchers at the University of Michigan and the University of Chicago analyzed data about immigration tweets from two political parties. Specifically, they used a computer program to review metaphors used in more than 400,000 messages by U.S. residents from 2018 to 2019.
Researchers selected seven concepts for the analysis: animal, vermin, parasite, physical pressure, water, commodity and war. Each one creates a distinct logic about the perceived threats and plausible remedies.
Metaphors related to each of these concepts may be blatant or implicit. For example, someone may overtly dehumanize immigrants by calling them "animals," but they could subtly evoke animal imagery by talking about "swarms" of people "flocking" across the border.
Dehumanizing metaphors can be used by people expressing strong opposition toward immigrants, such as in deploying the "water" metaphor when lamenting about immigrants "pouring" into the country, the study indicated. However, many metaphors are so commonplace in how we talk about immigrants that they are accepted as natural and neutral, such as in references to "waves" of immigration throughout U.S. history.
"When dehumanizing metaphors are used, it can increase discrimination and harsh policies," said study co-author Ceren Budak, associate professor at the U-M School of Information.
Julia Mendelsohn, postdoctoral scholar at the University of Chicago and lead author, said the way strong ideas influence how people use metaphors is different and a bit complicated in the two political parties.
People who are very conservative use more metaphors than those who are more in the middle. But for liberals, it's different. Liberals who are very far left don't use as many metaphors about water or buying and selling things, but they do use more metaphors about creatures, like animals.
The result is about the change in retweets, Mendelsohn said. When people use creature metaphors in tweets, they tend to get retweeted more, and this effect is stronger for liberals than conservatives.
"So, the difference between liberals' tweets with and without creature metaphors is greater than the difference between conservatives' tweets with and without creature metaphors," she said.
The researchers said their method used large language models, specifically ChatGPT, to identify metaphors. These models can be useful tools for in-depth analysis of language in political discussions, but careful study design and human evaluation of these models is crucial, Mendelsohn said.
